• The Journal of the Korea Contents Association
• /
• v.16 no.5
• /
• pp.251-258
• /
• 2016

Recently, the demand for fast storage devices is rapidly increasing in cloud platforms, social network services, etc. Despite the development of fast storage devices, the traditional Linux I/O stack is not able to exploit the full extent of the performance improvement since it has been optimized for disk-based storage devices. In this paper, we propose an optimized I/O stack which can fully utilize the I/O bandwidth and latency of fast storage devices. To this end, we design a new I/O interface to replace the current block I/O interface and optimize our I/O interface. Our optimized I/O interface bypasses operations/layers in block I/O subsystems of the current Linux I/O stack to fully exploit fast storage devices. We also optimize the Linux file systems such as ext2 and ext4 to run on our I/O interface. We evaluate our I/O stack with multiple benchmarks and the experimental results show that our I/O stack achieves 1.7 times better throughput compared to traditional Linux I/O stack.

• Journal of KIISE
• /
• v.43 no.1
• /
• pp.27-33
• /
• 2016

Virtualization has become one of the most helpful techniques in computing systems, and today it is prevalent in several computing environments including desktops, data-centers, and enterprises. However, since I/O layers are implemented to be oblivious to the I/O behaviors on virtual machines (VM), there still exists an I/O scalability issue in virtualized systems. In particular, when a multi-queue solid state drive (SSD) is used as a secondary storage, each system reveals a semantic gap that degrades the overall performance of the VM. This is due to two key problems, accelerated lock contentions and the I/O parallelism issue. In this paper, we propose a novel approach, including the design of virtual CPU (vCPU)-dedicated queues and I/O threads, which efficiently distributes the lock contentions and addresses the parallelism issue of Virtio-blk-data-plane in virtualized environments. Our approach is based on the above principle, which allocates a dedicated queue and an I/O thread for each vCPU to reduce the semantic gap. Our experimental results with various I/O traces clearly show that our design improves the I/O operations per second (IOPS) in virtualized environments by up to 155% over existing QEMU-based systems.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04b
• /
• pp.106-108
• /
• 2002

본 연구에서는 멀티미디어 데이터 검색에 클러스터링과 인덱싱 기법을 같이 적용하여 유사할 이미지끼리는 인접 디스크에 클러스터하고 이 클러스터에 접근하는 인덱스를 구축하여 검색이 빠르게 이루어지는 유사 검색방법을 제시한다. 이 연구에서는 트리 유사 구조의 인덱스 대신 해싱 방법을 이용하며 검색시 I/O시간을 줄이기 위해 오브젝트를 가진 클러스터 위치를 찾는데 한번의 I/O를 사용하고 이 클러스터를 읽기 위해 연속주인 파일 I/O를 사용하여 클러스터를 찾는 데용을 최소화한다 클러스터인덱싱 접근은 트리 유사 구조와 임의 I/O를 사용한 내용기반의 이미지 검색보다 효율적인 검색 적합성을 보이며 연속적 I/O를 통해 검색 미용을 낮춘다.

• Journal of the Korean Ceramic Society
• /
• v.31 no.8
• /
• pp.861-868
• /
• 1994

CaO-Al2O3 glass is a good candidate as optical fiber amplifier and laser. In this study, optical properties for 4I13/2longrightarrow4I15/2 transition of Er3+ ions doped in CaO-Al2O3 glasses were investigated. Optical absorptions, radiative transition probabilities and lifetimes for 4I13/2 level were evaluated by using Judd-Ofelt theory. Also, induced- emmision cross-sections of 4I13/2longrightarrow4I15/2 transition were calculated. Radiative transition probability and lifetime of 4I13/2 level were 144.6s-1, 690$mutextrm{s}$ respectively for 60CaO.40Al2O3 glass(FS0) and 152.6s-1, 660 $mutextrm{s}$ for 54 CaO.36Al2O3.10SiO2 glass (FS10). Each induced-emission cross sections for FS0 and FS10 was 0.749$\times$10-20 $\textrm{cm}^2$, 0.892$\times$10-20 $\textrm{cm}^2$. Obtained values were comparable with those of ZBLA glass studied as optical fiber amplifier and laser material.

• Journal of KIISE
• /
• v.41 no.9
• /
• pp.605-610
• /
• 2014

Now, virtualization is no more emerging research area, and we can easily find its application in our circumstance. Nevertheless, I/O workloads are reluctant to be applied in virtual environment since they still suffer from unacceptable performance degradation due to virtualization latency. Many previous papers identified that virtual I/O overhead is mainly caused by exits and redundant I/O stack, and proposed several techniques to reduce them. However, they still have some limitations. In this paper, we introduce a novel I/O virtualization framework which improves I/O performance by exploiting multicore architecture. We applied our framework to the virtual network, and it improves TCP throughput up to 169%, and decreases UDP latency up to 38% on the network with the 10Gbps NIC.

• ETRI Journal
• /
• v.23 no.4
• /
• pp.151-157
• /
• 2001

Spectroscopic and thermal analysis indicates that tellurite glasses doped with $B_2O_3$ and $GeO_2$ are promising candidate host materials for wide-band erbium doped fiber amplifier (EDFA) with a high 980 nm pump efficiency. In this study, we measured the thermal stabilities and the emission cross-sections for $Er^{3+}:^{4}I_{13/2}\;{\to}\;^{4}I_{15/2}$ transition in this tellurite glass system. We also determined the Judd-Ofelt parameters and calculated the radiative transition rates and the multiphonon relaxation rates in this glass system. The 15 mol% substitution of $B_2O_3$ for $TeO_2$ in the $Er^{3+}-doped\;75TeO_2-20ZnO-5K_2O$ glass raised the multiphonon relaxation rate for $^4I_{11/2}\;{\to}\;^4I_{13/2}$ transition from 4960 $s^{-1}$ to 24700 $s^{-1}$, but shortened the lifetime of the $^4I_{13/2}$ level by 14 % and reduced the emission cross-section for the $^4I_{13/2}\;{\to}\;^4I_{15/2}$ transition by 11%. The 15 mol% $GeO_2$ substitution in the same glass system also reduced the emission cross-section but increased the lifetime by 7%. However, the multiphonon relaxation rate for $^4I_{11/2}{\to}^4I_{13/2}$ transition was raised merely by 1000 $s^{-1}$. Therefore, a mixed substitution of $B_2O_3$ and $GeO_2$ for $TeO_2$ was concluded to be suitable for the 980 nm pump efficiency and the fluorescence efficiency of $^4I_{13/2}{\to}^4I_{15/2}$ transition in $Er^{3+}-doped$ tellurite glasses.

• Journal of the Korean Chemical Society
• /
• v.39 no.3
• /
• pp.163-175
• /
• 1995

Metal free phthalocyanine($H_2Pc$) partially doped with iodine, $H_2Pc(I)x$, has been made to improve photosensitizing efficiency of ZnO/$H_2Pc$. The content of iodine dopant level(x) for $H_2Pc(I)x$ upon $H_2Pc$ polymorphs was characterized as ${\chi}-H_2Pc(I)_{0.92}$ and ${\beta}-H_2Pc(I)_{0.96}$ by elemental analysis. Characterization of iodine-oxidized $H_2Pc$ were investigated by TGA (thermogravimetric analysis), UV-Vis, FT-IR, Raman and ESR (electron spin resonance) spectrum, and the adsorption properties of $H_2Pc(I)x$ on ZnO were characterized by means of Raman and ESR studies. TGA for $H_2Pc(I)x$ showed a complete loss of iodine at approximately 265$^{\circ}C$ and the Raman spectrum of $H_2Pc(I)x$ and ZnO/$H_2Pc(I)x$ at 514.5 nm showed characteristic $I_3^-$ patterns in the frequency region 90∼550 $cm^{-1}$. ZnO/$H_2Pc(I)x$ exhibited a very intense and narrow ESR signal at $g=2.0025{\pm}0.0005$ compared to $H_2Pc$/ZnO. Iodine doped ZnO/$H_2Pc(I)x$ showed a better photosensitivity compared to iodine undoped ZnO/$H_2Pc$. That is, the surface photovoltage of ${\chi}-H_2Pc(I)_{0.92}$/ZnO was approximately 31 times greater than that of ZnO/${\chi}-H_2Pc$ and ZnO/${\beta}-H_2Pc(I)_{0.96}$ was 5 times more efficient than ZnO/${\beta}-H_2Pc$ at 670 nm. And the dependence of photosensitizing effect upon $H_2Pc$ polymorphs was exhibited that the surface photovoltage of ZnO/${\chi}-H_2Pc(I)_{0.92}$ was approximately 5 times greater than ZnO/${\beta}-H_2Pc(I)_{0.96}$ at 670 nm. Therefore Iodine doping of H_2Pc$ resulted in increase in photoconductivity of $H_2Pc$ and photovoltaic effect of ZnO/$H_2Pc$ in the visible region.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.55 no.5
• /
• pp.229-235
• /
• 2006

High speed PC-based DIO(Digital I/O) system that consists of a master device and slave I/O devices is developed. The PCI interfaced master device controls all of serial communications, reducing the load on the CPU to a minimum. The slave device is connected from the master device and another slave device is connected to the slave device, it can repeated to maximum 64 slave devices. The slave device has 3 types I/O mode, such as 16 bits input-only, 16 bits output-only, and 8bits input-output. The master device has 2 rings which can take 64 slaves each. Therefore, total I/O points covered by the master is 2048 points. The slave features 3 types of input/output function interchangeability by DIP switch settings. Library, application, and device driver software for the DIO system that have a secure and a convenient functionality are developed.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.10 no.3
• /
• pp.171-178
• /
• 2015

Recently NAND flash memory has been found to be the primary cause of low performance in the smart device. NAND flash memory is different from each other the execution time of I/O operations that flash file system is required. Therefore, it is necessary to compare and analyze the flash file system I/O performance for the efficient I/O on smart device. In this paper, it was tested and analyzing the I/O performance of the YAFFS2, JFFS2, UBIFS. Experimental results most read I/O performance is good, but the writing I/O performance is not good. For UBIFS, showed a more good I/O performance compared to other flash file system.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10a
• /
• pp.657-660
• /
• 2000

병렬 파일 시스템은 늦은 디스크 I/O로 인한 성능 저하를 개선하기 위해 병렬 I/O를 제공한다. 이때 계산과 디스크 I/O를 중첩시키는 선반입 기법으로 디스크 I/O로 인한 성능 저하를 더욱 개선할 수 있다. 하지만 I/O 위주의 프로그램에서는 선반입으로 인하여 시스템에서 제공하는 I/O 대역폭을 넘어 최악의 경우 기존의 선반입 기법은 성능개선을 위한 최선이 될 수 없을 뿐 아니라 선반입 기법 자체가 과부하가 될 수 있다. 본 논문에서는 이런 상황을 고려하여 I/O 대역폭 개선을 위한 이단 선반입 기법을 제시하여 성능개선을 제공한다.